Processor/memory module with foldable substrate
A packaging approach reduces the overall footprint for interconnecting multiple semiconductor devices. In an embodiment, a processor mounts onto the center of a substrate with flexible appendages and memory components mount to the flexible appendages. The appendages fold over the processor to produce a processor/memory module. The processor/memory module occupies less area on the main printed circuit board than the laterally interconnected processor and memory devices would occupy.
This application is a continuation of U.S. patent application Ser. No. 10/845,373 filed May 13, 2004, entitled “PROCESSOR/MEMORY MODULE WITH FOLDABLE SUBSTRATE,” which is hereby incorporated by reference and which claims the benefit of U.S. Provisional Application No. 60/471,544, filed May 19, 2003, entitled “PROCESSOR/MEMORY MODULE WITH FOLDABLE SUBSTRATE”, the entirety of which is hereby incorporated herein by reference.
BACKGROUND1. Field of Invention
The invention relates to increasing the density of integrated circuits on a printed circuit board, and more particularly, to increasing the density of memory associated with a processor on a printed circuit board.
2. Description of Related Art
Modern electronic devices, such as computers and the like, typically include semiconductor devices, such as integrated circuits. Integrated circuits are microcircuits formed on a semiconductor substrate and packaged in a ceramic, plastic or epoxy package having multiple external terminals. The microcircuits are wire-bonded within the package to the external terminals. Integrated circuits are commonly found in dual-in-line packages (DIPs), for example, in which the integrated circuit is housed in a rectangular casing with two rows of connecting pins on either side.
When the terminals of the packages are connected to a printed circuit board, the integrated circuits are electrically connected to other integrated circuits and electrical components through or by way of traces on the printed circuit board to form system level electronic circuits.
As the size and complexity of semiconductors increases, packaging schemes have become bulky in providing routing channels for the high number of signals that are routed from the integrated circuit to the package pins. Semiconductor manufacturers have developed several packaging schemes that achieve a smaller package footprint as compared to conventional dual-in-line packages. One such example is a single-in-line package (SIP). In the single-in-line package, the connecting pins protrude from one side of the housing, allowing denser packaging of the semiconductor devices on the printed circuit board.
Additionally, as modern electronic devices are driven to ever increasing functionality and decreasing size, the printed circuit boards within the electronic devices are driven to increased integrated circuit densities. The desire to provide the capability of integrated circuits to be used in relatively small devices limits the extent to which multiple integrated circuits can be laterally interconnected while still fitting within the device.
Lateral extension and interconnection of semiconductor devices tends to lead to relatively long interconnects or traces between devices which increases the signal propagation delay, and thus, decreases the circuit operating speed. Further, lengthy traces increase both the radio-frequency interference (RFI) and electromagnetic interference (EMI) emitted from the printed circuit board.
To avoid the limitations of laterally interconnecting integrated circuits, some manufacturers mount the integrated circuits on a substrate, which is folded onto itself and secured to form a subassembly. The subassembly is then mounted on the main printed circuit board, where the subassembly increases the integrated circuit density of the main printed circuit board by occupying a smaller area than the laterally interconnected semiconductors would occupy.
However, the semiconductors forming the subassembly generate heat, which is often difficult to remove from the densely populated and compact subassembly. If the excess heat is not removed from the subassembly, the semiconductors may not function properly and may ultimately fail.
Further, it is often difficult to test some pins or interconnect nodes on the densely populated subassembly because the pins or interconnect nodes are buried in the folded subassembly. It may also not be possible to access the pins with test equipment. Even if the nodes are not buried, the probes on the test equipment may be too large or bulky to adequately reach areas on the subassembly.
SUMMARY OF THE INVENTIONThe invention relates to packaging a processor and its associated memory onto a subassembly, which mounts onto a main printed circuit board. The subassembly saves space on the main printed circuit board.
In one embodiment, an integrated circuit subassembly that provides for removal of heat from the subassembly increases circuit density on a printed circuit board.
In another embodiment, an integrated circuit subassembly that provides for testing of the subassembly increases circuit density on a printed circuit board.
Another advantage is shorter trace lengths between the processor and the memory input and output signals (I/O's). Thus, increasing the circuit speed, and reducing RFI and EMI emissions.
The subassembly optionally includes a heatsink to remove heat from the processor and memory components.
In an embodiment, the subassembly comprises a printed circuit board comprising a center section, and flexible appendages or wings. A processor mounts in the center section of the printed circuit board along with an optional heat sink. The memory components mount on the flexible appendages of the subassembly. The memory components mount on either the top surface of the appendages or the bottom surface of the appendages. In another embodiment, the memory components mount on both on the top and the bottom surfaces of the appendages.
In an embodiment, the wings fold over the center section such that the first wing folds over the center section, the second wing folds over the first wing, the third wing folds over the second wing, and the fourth wing folds over the third wing. The wings are then secured. In an embodiment, a ball grid array connects the processor/memory assembly to the main board.
In another embodiment, a heatsink mounts onto the top of the processor. The wings fold up to meet the heatsink and are then secured. The heatsink transfers heat away from the processor and the memory components.
In another embodiment, the processor/memory module comprises a printed circuit board comprising a center section, and flexible appendages comprising connectors or sockets. A processor mounts onto the center section of the printed circuit board and a heatsink comprising fins mounts onto the processor. Memory subassemblies, such as, for example, single-in-line memory modules (SIMMs), connect to the connectors, such as, for example, in-line connectors. When the flexible appendages fold upward, the memory subassemblies form a stacked arrangement. The heatsink fins support the stacked arrangement of memory modules and help dissipate at least a portion of the heat generated by the processor and memory components. In an embodiment, the connectors or sockets mount to the top surface or the bottom surface of the flexible appendages. In another embodiment, the connectors or sockets mount to the top and the bottom surfaces of the flexible appendages.
In another embodiment, the processor/memory subassembly comprises a heatsink, and a printed circuit board comprising a center section, and at least one flexible appendage or wing. The heatsink comprises a center and two end sections. A processor mounts onto the center section of the printed circuit board and the center section of the heatsink mounts onto the top of the processor. The flexible wing folds over the processor and the heatsink such that the center section of the heat sink is interposed between the processor and the flexible wing. The wing comprises connectors or sockets positioned on the wing surface. Memory subassemblies, such as, for example, single in-line memory modules (SIMMs), connect to the connectors, such as, for example, in-line connectors, such that the memory subassemblies form a parallel arrangement above the processor. The end sections of the heatsink further comprise fins to transfer at least a portion of the heat generated by the processor and the memory components away from the processor and memory components.
A further embodiment relates to properly terminating the input/output lines between the processor and the memory through impedance matching. Proper termination of the lines, in addition to shorter traces between the processor and the memory, reduces crosstalk between the traces and reduces electromagnetic radiation (EMI) of the traces.
Yet another embodiment of the invention relates to correcting problems on the processor/memory subassembly that are not easily corrected on the main board. For example, the subassembly can be used to correct missing or incorrect traces on the main board.
An additional embodiment of the invention relates to using the processor/memory module to test signals not available on the main board. The flexible wing comprises connectors that connect to test equipment through test cables. The connectors are spaced and dimensioned to facilitate testing. In an embodiment, the processor/memory module permits testing of the memory components. In another embodiment, the module permits the testing of signals and components from the main printed circuit board.
For purposes of summarizing the invention, certain aspects, advantages and novel features of the invention have been described herein. It is to be understood that not necessarily all such advantages may be achieved in accordance with any particular embodiment of the invention. Thus, the invention may be embodied or carried out in a manner that achieves or optimizes one advantage or group of advantages as taught herein without necessarily achieving other advantages as may be taught or suggested herein.
These and other objects and advantages of the present invention will be more apparent from the following description taken in conjunction with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGSA general architecture that implements the various features of the invention will now be described with reference to the drawings. The drawings and the associated descriptions are provided to illustrate embodiments of the invention and not to limit the scope of the invention. Throughout the drawings, reference numbers are re-used to indicate correspondence between referenced elements.
For a more detailed understanding of the invention, reference is first made to
In this example, the subassembly 100 has four wings for illustrative purposes only. In other embodiments, the subassembly 100 can have less than four wings 104 or more than four wings 104. It is understood that the number of appendages or wings 104 is not limited to four and the apparatus and methods described below apply to subassemblies 100 with more or less than four appendages 104.
In the illustrated example of
In an embodiment, the center section 102 comprises a rigid printed wiring board (PWB) made of multiple layers of epoxy glass laminate, as is well know in the art. In another embodiment, the center section 102 comprises, for example, a substrate composed of aluminum oxide, a substrate composed of aluminum nitride, a polyimide flex board, or the like.
Referring in particular to
In an embodiment, the appendages 104 comprise polyimide flex boards, or other flexible circuit boards as are known in the art. The flexible appendage 104 allows the appendage 104 to fold up towards a top surface of the center section 102. In another embodiment, the flexible appendage 104 allows the appendage 104 to fold over the top surface of the center section 102.
Referring in particular to
The flexible appendages 104 can connect to the center section 102, for example, through soldering, solder bumps, electrically conductive adhesive dots, pins, and the like.
In another embodiment, the appendages 104 comprise rigid printed wiring boards (PWBs) made of multiple layers of epoxy glass laminate, substrates composed of aluminum oxide, substrates composed of aluminum nitride, or other well known printed wiring board materials. Each rigid appendage 104 further comprises a flexible interconnect 106 between the appendage 104 and the center section 102 such that the connection between the appendage 104 and the center section 102 is foldable. Examples of a flexible interconnect material are polyimide flex board, and the like. It is understood that the invention is not limited to the materials described.
Referring in particular to
In an embodiment, an edge of the flexible interconnect section 106 interposes between layers of the rigid appendage 104, and an opposite edge of the interconnect section 106 interposes between the layers of the rigid center section 102. In other embodiments, the flexible interconnect 106 attaches to the rigid wing 104 and the center section 102 by a wide variety of other methods, such as, for example, soldering, solder bumps, electrically conductive adhesive dots, pins, and the like.
The flexible interconnect section 106 allows the appendage 104 to fold, such as, toward the top surface of the center section 102. In another embodiment, the flexible interconnect 106 permits the appendage 104 to fold over the top surface of the center section 102.
Subassembly 100 further comprises a first semiconductor device 110 mounted onto the top surface of the center section 102. The semiconductor device 110 electrically connects to the center section 102 such that at least one signal from the semiconductor device 110 conducts through traces on the center section 102 to the flexible appendage 104 and/or to the main printed circuit board.
The first semiconductor device 110 comprises an integrated circuit as is well known in the art. In an embodiment, the semiconductor device 110 is a processor. In other embodiments, the processors can comprise controller circuitry, processor circuitry, processors, general purpose single-chip or multi-chip microprocessors, digital signal processors, embedded microprocessors, microcontrollers and the like.
In an embodiment, the semiconductor device 110 is a bare die attached to the center section 102 through solder bumps. A mask applied to the top surface of the center section 102 defines sites for an array of solder bumps in preparation for attaching the semiconductor device 110 to the subassembly 100. The semiconductor device 110 comprises bumps, such as nickel-gold bumps or solder on a surface of the semiconductor die for mating with the array of solder bumps or solderable pads on the center section 102. A reflow process is then used to attach the semiconductor device 110 to the subassembly 100. It is understood that the invention is not limited to the particular solder materials described.
In another embodiment of the invention, the semiconductor device 110 connects to the center section 102 using wire bond edge connections.
In another embodiment, the semiconductor device 110 can be packaged as a dual-in-line package, a single-in-line package, a ball grid array, a pin grid array, a leadless chip, a flip-chip, or any other semiconductor package type as is well know in the art.
Preferably, the semiconductor device 110 connects to the subassembly 100 with high temperature solder.
Subassembly 100 further comprises at least one device 108. In an embodiment, each wing 104 comprises at least one device 108, which mounts onto at least one surface of appendage 104. In another embodiment, at least one device 108 mounts onto at least one surface of at least one appendage 104.
The device 108 electrically connects to the wing 104 such that at least one signal from the device 108 conducts through traces on the wing 104 to the center section 102 and/or the main printed circuit board. The appendages 104 provide electrical interconnects for connecting signals and power supply lines between the semiconductor device 110 and the devices 108.
In an embodiment, the device 108 comprises a connector or a socket, which electrically connects to the appendage 104.
In another embodiment, the device 108 comprises an integrated circuit as is well known in the art. In an embodiment, the device 108 comprises a memory device such as random access memory (RAM), static random access memory (SRAM), non-volatile random access memory (NVRAM), dynamic random access memory (DRAM), multi-bank dynamic random access memory (MDRAM), synchronous dynamic random access memory (SDRAM), magnetoresistive random access memory (MRAM), read only memory (ROM), programmable read only memory (PROM), erasable programmable read only memory (EPROM), electrically erasable programmable read only memory (EEPROM), flash memory, and the like.
In an embodiment, the semiconductor device 108 can be a bare die attached to the appendages 104 using solder applied to sites on the appendages 104. Preferably, a process with pre-bumped semiconductor device 108 uses a solder process while providing either a 95:5 Pb/Sn solder bump or an electroless Ni/Au bump. Other solder bumps, such as a 90:10 Pb/Sn solder bump, and the like can also be used. Preferably, the bump height is approximately 25-50 mm. It is understood that the invention is not limited to the particular solder materials described.
In another embodiment of the invention, the semiconductor devices 108 connect to the wings 104 using wire bond edge connections.
In another embodiment, the semiconductor device 108 can be packaged as a dual-in-line package, a single-in-line package, a ball grid array, a pin grid array, a leadless chip, a flip-chip, or any other semiconductor package type as is well know in the art.
Preferably, the devices 108 connect to subassembly 100 with high temperature solder.
In an embodiment, the semiconductor devices 108 comprise identical integrated circuits. In another embodiment, the semiconductor devices 108 comprise various integrated circuits.
The subassembly 100 further comprises a connecting device 200. The connecting device 200 electrically connects signals and power supply lines of the subassembly 100 with a main printed circuit board. In the embodiment shown in
In an embodiment, the module 300 has a footprint approximately the same size as the footprint of the semiconductor device 110. In another embodiment, the module 300 occupies substantially less area than the area occupied by laterally interconnecting the semiconductor device 110 and the devices 108 on the main printed circuit board.
Referring in particular to
In an embodiment of the invention, a non-conductive insulating film, such as polyimide film, for example, is applied to the bottom surface of the appendages 104a-d to provide electrical isolation between the folded layers of the module 300. Alternatively, a low flow non-conductive epoxy adhesive could be used to secure the folded appendages 104a-d above the center section 102. For applications requiring enhanced thermal conductivity, an adhesive with improved thermal properties comprising aluminum nitride, boron nitride or the like could be used.
In another embodiment, a thermal cap can reside over the center section 102 and foldable appendages 104a-d provides thermal heat transfer from the devices 110 and 108 to outside the module 300. The thermal cap additionally provides moisture protection for the devices 110 and 108, and electromagnetic interference (EMI) ground shielding.
In a further embodiment of the invention, the subassembly 100 comprises one appendage 104, which is folded over the center section 102 to form the module 300. In another embodiment of the invention, the subassembly 100 comprises less than four or more than four appendages 104, which are folded over to form the module 300.
In an embodiment of the invention, the semiconductor device 110 comprises a processor, the devices 108 comprise memory devices, and the module 300 comprises a processor/memory subassembly.
In this example, the subassembly 400 has two wings 410, 412 for illustrative purposes only. In other embodiments, the subassembly 400 can have less than two wings 410, 412 or more than two wings 410, 412. It is understood that the number of appendages or wings 410, 412 is not limited to two and the apparatus and methods described below apply to subassemblies 400 with more or less than two appendages 410, 412.
Referring in particular to
Module 400 further comprises a heatsink 414. The shape of the heatsink 414 is a rectangular solid. In other embodiments, the heatsink 414 can be, for example, a polyhedron such as a cube, a rectangular prism, a pyramid, or the like. In yet other embodiments, the heatsink 414 can be circular, such as, for example, a sphere, a cylinder, or the like.
Referring in particular to
Typically, heat sinks comprise a plurality of fins. The fins provide the heatsink with a greater surface area from which heats dissipates. The heatsink 414 comprises a base and a plurality of fins. In an embodiment, the fins form vertical planes placed perpendicular to and spaced longitudinally across the heatsink base.
Referring in particular to
In another embodiment, for example, the fins can form posts placed perpendicular to and spaced in an array pattern across the length and width of the base. The posts may be square, rectangular, polygonal, circular, or the like in cross section. In other embodiments, the structure of the heatsink 414 may be configured in a wide variety of manners, as required or desired.
In an embodiment, the heatsink 414 is extruded aluminum. Other typical heatsink materials include, for example, copper, copper coil, stainless steel, iron, metal alloys, or any other thermally conductive material. The heatsink 414 can be fabricated by a wide variety of methods, such as, for example, extrusion, stamping folding, drawing, molding, casting, forging, welding, bonding, punching CNC machining, or the like. The finish on the heatsink 414 can be chromated, anodized, powder coated, or the like. It is understood that the heatsink 414 is not limited to the particular materials, fabrication methods, and finishes described.
The heatsink 414 sets on the semiconductor device 110 and the appendages 410, 412 fold upward such that the devices 108 meet the heatsink 414. In an embodiment, the appendages 410, 412 and the center section 102 form approximately a right angle.
Preferably, a non-conductive thermal grease placed between the semiconductor device 110 and the heatsink 414 and/or between the devices 108 and the heatsink 414 provides greater thermal conductivity between the devices 110, 108 and the heatsink 414.
The subassembly 400 further comprises at least one clamp 416, which holds the wings 410 and 412 against the heatsink 414. In an embodiment, the clamp 416 is a metal strap, which clips over the sides of the appendages 410, 412 and rests on the fins of the heatsink 414. In an embodiment, the clamp 416 is fabricated from steel, aluminum, plastic, or the like. It is understood that the clamp 416 is not limited to the particular materials described.
Referring in particular to
In another embodiment, the module 400 comprises more than two or less than two foldable appendages which fold upward to meet the heatsink 414.
In an embodiment, the semiconductor device 110 is a processor, the devices 108 are memory components, and the module 400 is a processor/memory subassembly.
In this example, the subassembly 500 has two wings 510, 512 for illustrative purposes only. In other embodiments, the subassembly 500 can have less than two wings 510, 512 or more than two wings 510, 512. It is understood that the number of appendages or wings 510, 512 is not limited to two and the apparatus and methods described below apply to subassemblies 500 with more or less than two appendages 510, 512.
Referring in particular to
The flexible appendages 510, 512 further comprise a plurality of connectors or sockets 520 and a plurality of memory subassemblies 530.
The memory subassembly 530 comprises a printed wiring board, at least one memory device, and an edge connector, such as, for example, found in single in-line memory modules (SIMMs) or dual in-line memory modules (DIMMs).
In an embodiment, the memory device of the memory subassembly 530 comprises, for example, random access memory (RAM), static random access memory (SRAM), non-volatile random access memory (NVRAM), dynamic random access memory (DRAM), multi-bank dynamic random access memory (MDRAM), synchronous dynamic random access memory (SDRAM), magnetoresistive random access memory (MRAM), read only memory (ROM), programmable read only memory (PROM), erasable programmable read only memory (EPROM), electrically erasable programmable read only memory (EEPROM), flash memory, and the like.
The edge connectors on the memory subassemblies 530 connect to the connectors or sockets 520. The connectors or sockets 520 mount to the foldable appendages 510, 512 on the top and on the bottom surfaces of the foldable appendages 510, 512. In another embodiment, the connectors or sockets 520 mount to the top surfaces or the bottom surfaces of the foldable wings 510, 512. The connectors 520 electrically connect to the wings 510, 512 such that at least one signal from the memory subassembly 530 conducts through traces on the wings 510, 512 to the center section 102.
Referring in particular to
The module 500 further comprises a heatsink 550. The heatsink 550 comprises a base 552, a top section 554, a plurality of sides 556, a hollow rectangular center support 558, and a plurality of fins 560. The center support 558 sets between the base 552 and the top section 554, and includes fins 560. The fins 560 extend horizontally from the center support 558, toward the wings 510, 512, and parallel to the base 552. The sides 556 extend down from opposite edges of the top section 554 and also include fins 560. The fins 560 extend horizontally from the sides 556, toward the center section 558, and parallel to the base 552.
Referring in particular to
Again, referring in particular to
Preferably, the heatsink 550 is extruded aluminum. Other typical heatsink materials include, for example, copper, copper coil, stainless steel, iron, metal alloys, or any other thermally conductive material. The heatsink 550 can be fabricated by a wide variety of methods, such as, for example, extrusion, stamping folding, drawing, molding, casting, forging, welding, bonding, punching CNC machining, or the like. The finish on the heatsink 550 can be chromated, anodized, powder coated, or the like. It is understood that the heatsink 550 is not limited to the particular materials, fabrication methods, and finishes described.
The heatsink 550 mounts onto the semiconductor device 110. Preferably, a non-conductive thermal grease placed between the semiconductor device 110 and the heatsink 550 provides greater thermal conductivity between the semiconductor device 110 and the heatsink 550.
The foldable appendages 510, 512, comprising the memory subassemblies 530 and the connectors or sockets 520, fold upward with respect to the center section 102. The memory subassemblies 530 form a stacked arrangement within the module 500. The fins 560 on the heatsink 550 are placed such that when one edge of the memory subassembly 530 connects to the connector or socket 520, the fin 560 supports the opposite edge of the memory subassembly 530.
Preferably, a non-conductive thermal grease placed between the memory subassemblies 530 and the heatsink 550 provides greater thermal conductivity between the memory subassemblies 530 and the heatsink 550.
Again, referring in particular to
Again, referring in particular to
The heatsink 550 supports the memory subassemblies 530. In addition, the heatsink 550 transfers at least a portion of the heat generated by the semiconductor device 110 and the memory subassemblies 530 away from the semiconductor device 110 and the memory subassemblies 530.
In an embodiment, the semiconductor device 110 comprises a processor and the module 500 comprises a processor/memory subassembly. In an embodiment, the footprint of the processor/memory module 500 is significantly less than the area the laterally connected memory devices on the memory subassemblies 530 and processor 110 would occupy on the main printed circuit board.
In this example, the subassembly 600 has one wing 610 for illustrative purposes only. In other embodiments, the subassembly 600 can have more than one wing 610. It is understood that the number of appendages or wings 610 is not limited to one and the apparatus and methods described below apply to subassemblies 600 with more than one wing 610.
Referring in particular to
The flexible appendage 610 further comprises a plurality of connectors or sockets 620 and a plurality of memory subassemblies 630.
The memory subassembly 630 comprises a printed wiring board, a plurality of memory devices and an edge connector, such as, for example, found in single in-line memory modules (SIMMs) or dual in-line memory modules (DIMMs).
In an embodiment, the memory subassembly 630 comprises at least one memory device, such as, for example, random access memory (RAM), static random access memory (SRAM), non-volatile random access memory (NVRAM), dynamic random access memory (DRAM), multi-bank dynamic random access memory (MDRAM), synchronous dynamic random access memory (SDRAM), magnetoresistive random access memory (MRAM), read only memory (ROM), programmable read only memory (PROM), erasable programmable read only memory (EPROM), electrically erasable programmable read only memory (EEPROM), flash memory, and the like.
Referring in particular to
The edge connectors on the memory subassemblies 630 connect to the connectors 620. The connectors or sockets 620 mount to the foldable appendage 610 on the bottom surface of the foldable appendage 610. In another embodiment, the connectors or sockets 620 mount to the top surface or both the top and bottom surfaces of the flexible wing 610. The connectors 620 electrically connect to the wing 610 such that at least one signal from the memory subassembly 630 conducts through traces on the wing 610 to the center section 102.
Referring in particular to
The module 600 further comprises a heatsink 650. The heatsink 650 comprises a base, and a plurality of fins. In an embodiment, the fins form vertical planes placed perpendicular to and spaced longitudinally across the heatsink base. The fins are located on either side of the heatsink base, leaving a center area of the heatsink 650 available to support the folded wing 610.
Referring in particular to
Again, referring in particular to
Again referring in particular to
Preferably, the heatsink 650 is extruded aluminum. Other typical heatsink materials include, for example, copper, copper coil, stainless steel, iron, metal alloys, or any other thermally conductive material. The heatsink 650 can be fabricated by a wide variety of methods, such as, for example, extrusion, stamping folding, drawing, molding, casting, forging, welding, bonding, punching CNC machining, or the like. The finish on the heatsink 650 can be chromated, anodized, powder coated, or the like. It is understood that the heatsink 650 is not limited to the particular materials, fabrication methods, and finishes described.
The center section of the heatsink 650 mounts to an upper surface of the semiconductor device 110. The flexible appendage 610, comprising the connectors or sockets 620 and the memory subassemblies 630, folds over the center of the heatsink 650 such that the heatsink 650 interposes between the flexible appendage 610 and the semiconductor device 110.
Preferably, a non-conductive thermal grease placed between the semiconductor device 110 and the heatsink 650 and/or between the flexible wing 610 and the heatsink 650 provides greater thermal conductivity between the devices 110, 620, and the heatsink 650.
Again referring in particular to
The heatsink 650 supports the memory subassemblies 630. In addition, the heatsink 650 conducts at least a portion of the heat generated by the semiconductor device 110 and the memory subassemblies 630 away from the semiconductor device 110 and the memory subassemblies 630.
In an embodiment, the semiconductor device 110 comprises a processor and the module 600 comprises a processor/memory subassembly. When the module 600 mounts to the main printed circuit board, the area of the main printed circuit board below the fins of the subassembly 600 is available to receive additional integrated circuits. The area of the main printed circuit board that the subassembly occupies is substantially less than the area the laterally connected processor 110 and the memory devices on the memory subassemblies 630 would occupy.
In this example, the subassembly 700 has one wing 710 for illustrative purposes only. In other embodiments, the subassembly 700 can have more than one wing 710. It is understood that the number of appendages or wings 710 is not limited to one and the apparatus and methods described below apply to subassemblies 700 with more than one appendage 710.
Referring in particular to
The flexible appendage 710 further comprises at least one connector 720, which can connect to test equipment through test equipment cables 730. The center section 102 may optionally comprise semiconductor devices 108, 110, which mount to the center section as described above.
Again, referring in particular to
In a further embodiment of the invention, connectors 720 can mount to a top surface, a bottom surface or both the top and bottom surfaces of the at least one flexible appendage 710. The connectors 720 electrically connect to the wing 710 such that al least one signal from the wing 710, the center section 102, and/or the main printed circuit board is available at the connector 720.
The test cables 730 connect to the connectors 720. The test cables 730 can connect to a variety of test equipment. Examples of test equipment are logic analyzers, harmonic analyzers, oscilloscopes, function generators, multimeters, and the like.
Timing, signal levels, impedance, and the like, of the semiconductor devices on the subassembly 700 and on the main printed circuit board may be monitored or tested through the module 700. Further, the module 700 may allow access to signals not available for testing on the main printed circuit board.
The test subassembly 700 can be folded and secured as illustrated in
The folded subassembly packaging approach described herein reduces the overall area for interconnecting multiple semiconductor devices on the main printed circuit board. In an embodiment of the invention, the footprint of the modules 300, 400, 500, 600, 700 is approximately the same as the footprint of the semiconductor device 110. In another embodiment of the invention, the modules 300, 400, 500, 600, and 700 occupy less area on the main printed circuit board than the devices 110, 108 and/or connectors 520, 620, 720 laterally connected on the main circuit board would occupy.
In an embodiment of the invention, the interconnecting traces between the modules 100, 200, 300, 400, 500, 600, and 700 and the main circuit board are properly terminated through impedance matching. Proper termination of the lines, in addition to shorter traces between the semiconductor devices, reduces crosstalk between the traces and reduces electromagnetic radiation (EMI) of the traces.
Another embodiment of the invention allows the modules 100, 200, 300, 400, 500, 600, and 700 to correct errors that were not corrected on the main board. For example, the module 100, 200, 300, 400, 500, 600, and 700 can be designed to correct missing or incorrect traces on the main board. Thus, reducing the complexity and cost of the main printed circuit board.
While certain embodiments of the inventions have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions. Indeed, the novel methods and systems described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the methods and systems described herein may be made without departing from the spirit of the inventions. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of the inventions.
Claims
1. A memory module comprising:
- a center section;
- a first flexible wing electrically connected to a first edge of the center section;
- a semiconductor device mounted onto the center section; and
- a plurality of integrated circuit modules mounted onto the first flexible wing, wherein the first flexible wing folds over the center section creating a first stacked arrangement of integrated circuit modules.
2. The memory module of claim 1 further comprising a second flexible wing electronically connected to a second edge of the center section, wherein the second flexible wing folds over the first stacked arrangement creating a second stacked arrangement of integrated circuit modules.
3. The memory module of claim 2 further comprising a clamp to hold the second stacked arrangement of integrated circuit modules.
4. The memory module of claim 1 further comprising a heatsink wherein the heatsink supports the first stacked arrangement, and wherein the heatsink transfers at least a portion of the heat generated by the first stacked arrangement away from the first stacked arrangement.
5. The memory module of claim 1 wherein the first flexible wing comprises a rigid section and a flexible interconnect.
6. The memory module of claim 1 wherein the semiconductor device is a processor.
7. The memory module of claim 1 wherein the integrated circuit modules comprise flash memory.
8. The memory module of claim 1 wherein integrated circuit modules comprise random access memory.
9. The memory module of claim 1 wherein the first flexible wing comprises a first surface and a second surface, wherein the first surface and the second surface are on opposite sides of the first flexible wing, and wherein the integrated circuit modules are mounted on the first surface and the second surface.
10. A method of increasing the density of integrated circuits on a main assembly comprising:
- providing a subassembly comprising a center section;
- providing a first flexible wing electrically connected to a first edge of the center section, wherein the first flexible wing comprises a first surface and a second surface, and wherein the first surface and the second surface are on opposite sides of the first flexible wing;
- populating the center section with a semiconductor device;
- populating the first flexible wing with a plurality of semiconductor modules; and
- folding the first flexible wing over the center section, wherein the semiconductor modules form a first stack within the subassembly.
11. The method of claim 10 further comprising supporting the first stack with a heatsink.
12. The method of claim 10 further comprising connecting the subassembly to a main assembly.
13. The method of claim 10 wherein populating the first flexible wing with the plurality of semiconductor modules comprises mounting semiconductor modules on the first surface and the second surface.
14. The method of claim 10 wherein the semiconductor modules comprise flash memory.
15. The method of claim 10 semiconductor device is a processor.
16. The method of claim 10 further comprising:
- providing a second flexible wing electronically connected to a second edge of the center section; and
- folding the second flexible wing over the first stack creating a second stack within the assembly.
17. The method of claim 16 further comprising providing a clamp to hold the second stack within the assembly.
18. The method of claim 10 further comprising providing a heatsink, wherein the heatsink supports the first stack within the assembly.
19. A subassembly comprising:
- a center section;
- means to electronically connect a flexible wing to an edge of the center section;
- means to mount a semiconductor device onto the center section;
- means to mount a plurality of integrated circuit modules to the flexible wing, wherein the integrated circuit modules form a stacked arrangement when the flexible wing folds over the center section; and
- a means to support the stacked arrangement and to transfer at least a portion of the heat generated in the stacked arrangement away from the stacked arrangement.
20. The subassembly of claim 19 wherein the flexible wing comprises a first surface and a second surface, wherein the first surface and the second surface are on opposite sides of the flexible wing, and further comprising means for mounting the integrated circuit modules on the first surface and the second surface.
Type: Application
Filed: May 23, 2006
Publication Date: Sep 21, 2006
Inventor: Mark Moshayedi (Newport Coast, CA)
Application Number: 11/439,617
International Classification: H05K 7/20 (20060101);